In Trojanowski's vision, doctors may eventually be able to screen for pre-Alzheimer's disease and treat it with drugs the same way they can screen patients for high blood pressure or cholesterol and prevent heart attacks and strokes.
A paper published yesterday in the Archives of Neurology compiled a number of different studies of the test, which requires a lumbar puncture.
The test measures relative concentrations of two key proteins, both of them connected with the abnormal brain deposits that characterize the disease. What separates the positive from negative results is a telltale pattern - an Alzheimer's signature.
One of the more striking results came from a study of 57 people who were initially diagnosed with what's called mild cognitive dementia - a decline in memory and thinking that can be clearly measured by doctors - and then went on to develop Alzheimer's disease within five years.
All of those people had shown an "Alzheimer's signature" on the test before they developed the full-blown disease, Trojanowski said.
Already, Bristol-Myers Squibb is using the screening test to identify subjects for a clinical trial to evaluate the effectiveness of a new drug to treat the disease at a very early stage, said a spokeswoman for the company.
For years, Alzheimer's disease was difficult to distinguish from other types of dementia until the patient died and characteristic brain deposits could be seen on autopsy, said Trojanowski. But when they gave this new screening test to Alzheimer's patients who subsequently died, it proved to be 97 percent accurate in predicting whose diagnosis would be confirmed postmortem.
The scientists don't yet know what the test can predict for those without obvious memory problems. Out of 114 healthy controls, 39 percent tested positive for the Alzheimer's signature.
Following them will reveal whether they carry an elevated risk. "We don't know, but the speculation is that these people are on a trajectory toward Alzheimer's disease," said Trojanowski.
All this work on telltale proteins is starting to confirm what many have suspected for a long time - that the disease creeps up on people for 10 or even 15 years before they get symptoms.
Over that time, abnormal deposits known as amyloid plaques and tangles slowly take root and grow in the brain, said neurologist Paul Aisen of the University of California San Diego.
For a while, many people can still function. "The brain is resilient," he said. "To different degrees in different people, the brain is able to compensate for a significant pathology."
The current thinking is that the disease is driven by the accumulation of protein deposits called beta-amyloid plaques, and drugs now in development are aimed at fighting these.
A good case can be made for getting people with these deposits into clinical trials of these drugs at a much earlier stage, said Aisen, and eventually perhaps even in people without any detectable symptoms if their cerebrospinal fluid tests positive.
Of the two proteins in the screening test, one is associated with the amyloid plaques - a fragment called beta-amyloid 1-42.
Counterintuitively, the concentration of beta-amyloid 1-42 is abnormally low in the cerebrospinal fluid of people with Alzheimer's disease, Aisen said, because so much of the protein is stuck together in clumps in the brain that there's less in circulation. The test also measures two forms of a protein called Tau that's associated with the tangles.
The test is unlikely to be widely used until there's something people can do to stave off the disease, said Jason Karlawish, a professor of medicine and ethicist at Penn.
Before scientists developed a test for Huntington's disease, which is deadly and incurable, many people with a family history said they'd want to know their fate. But when the test was finally developed, very few went ahead and got it, he said.
But the Alzheimer's test perhaps might lead to new treatments by helping get the right patients into clinical trials.
Today, patients with mild but measurable symptoms are often diagnosed with mild cognitive impairment because doctors can't be sure of the root cause, said David Holtzman, an Alzheimer's disease researcher at Washington University in St. Louis.
Strokes, depression, brain tumors, or other diseases can sometimes cause mild cognitive impairment, he said. With this new screening test, doctors are much better equipped to sort out those who have early Alzheimer's.
Since there's still no cure, it's unlikely doctors will recommend this test except as part of scientific research, said University of California's Aisen. "That will change if we get a treatment that's effective. . . . Then we're going to be screening everybody when they reach a certain age."
Contact staff writer Faye Flam at 215-854-4977 or firstname.lastname@example.org.